Comparative Analysis of Denitrifying Activities of Hyphomicrobiumnitrativorans, Hyphomicrobium denitrificans, and Hyphomicrobiumzavarzinii

Christine Martineau,* Florian Mauffrey, Richard Villemur

INRS-Institut Armand-Frappier, Laval, QC, Canada

Hyphomicrobium spp. are commonly identified as major players in denitrification systems supplied with methanol as a carbonsource. However, denitrifying Hyphomicrobium species are poorly characterized, and very few studies have provided informa-tion on the genetic and physiological aspects of denitrification in pure cultures of these bacteria. This is a comparative study ofthree denitrifying Hyphomicrobium species, H. denitrificans ATCC 51888, H. zavarzinii ZV622, and a newly described species,H. nitrativorans NL23, which was isolated from a denitrification system treating seawater. Whole-genome sequence analysesrevealed that although they share numerous orthologous genes, these three species differ greatly in their nitrate reductases, withgene clusters encoding a periplasmic nitrate reductase (Nap) in H. nitrativorans, a membrane-bound nitrate reductase (Nar) inH. denitrificans, and one Nap and two Nar enzymes in H. zavarzinii. Concurrently with these differences observed at the geneticlevel, important differences in the denitrification capacities of these Hyphomicrobium species were determined. H. nitrativoransgrew and denitrified at higher nitrate and NaCl concentrations than did the two other species, without significant nitrite accu-mulation. Significant increases in the relative gene expression levels of the nitrate (napA) and nitrite (nirK) reductase genes werealso noted for H. nitrativorans at higher nitrate and NaCl concentrations. Oxygen was also found to be a strong regulator ofdenitrification gene expression in both H. nitrativorans and H. zavarzinii, although individual genes responded differently inthese two species. Taken together, the results presented in this study highlight the potential of H. nitrativorans as an efficientand adaptable bacterium that is able to perform complete denitrification under various conditions.

Hyphomicrobium spp. are restricted facultative methylotrophsthat reproduce by budding at the tip of a polar prostheca.

They are ubiquitous in water and soil but can also be found insewage treatment plants (1). Some strains have been characterizedby their denitrification capacities (2–5). They have often beenidentified as major players in denitrification systems supple-mented with methanol (4, 6–8), and their presence has been asso-ciated with high denitrification rates (9, 10).

Denitrification takes place in bacterial cells where N-oxidesand/or N-oxyanions serve as the terminal electron acceptor in-stead of oxygen (O2) for energy production when oxygen deple-tion occurs, leading to the production of gaseous nitrogen (N2)(11). Four sequential reactions are essential for the reduction ofnitrate to gaseous nitrogen via nitrite, nitric oxide, and nitrousoxide intermediates, and each of these reactions is catalyzed bydifferent enzymes, namely, nitrate reductases (Nar and Nap), ni-trite reductase (Nir), nitric oxide reductase (Nor), and nitrousoxide reductase (Nos) (12–14).

Previous studies using DNA hybridization with probes target-ing genes encoding the catalytic subunit of denitrification reduc-tases (narG, nirK, nirS, and nosZ) detected denitrification genes inseveral members of the genus Hyphomicrobium, including H.denitrificans and H. zavarzinii (2, 3). H. denitrificans (5) is com-monly found in soils, freshwater environments, and activatedsludge. This bacterium has the capacity to denitrify (2), and itsnitrite and nitrous oxide reductases were previously characterized(15–18). H. zavarzinii is a soil bacterium that has been studiedmainly for the biotechnological potential of the exceptional form-aldehyde dehydrogenase found in strain ZV580 (19, 20). Recently,we described a new denitrifying species of the genus Hyphomicro-bium, H. nitrativorans NL23T (21, 22). This bacterium was isolated

from a methanol-fed denitrification system treating seawater atthe Montreal Biodome in Canada. H. nitrativorans was one of themost abundant bacterial species in the denitrifying biofilm, alongwith Methylophaga nitratireducenticrescens JAM1T, which was alsoisolated from this biofilm and which can reduce nitrate into nitrite(23, 24). Given that pure cultures of H. nitrativorans grow poorlyin seawater (21), we believe that M. nitratireducenticrescens JAM1and H. nitrativorans NL23 work together in syntrophy in the bio-film to achieve the complete denitrification pathway.

Despite the presence of denitrifying Hyphomicrobium spp. inmany environments or denitrification processes, genetic andphysiological information regarding their denitrification capaci-ties is poorly documented. This could be related to the fact thatHyphomicrobium spp. can be hard to culture, especially underdenitrifying conditions. In this study, three denitrifying Hyphomi-

Received 17 March 2015 Accepted 12 May 2015

Accepted manuscript posted online 15 May 2015

Citation Martineau C, Mauffrey F, Villemur R. 2015. Comparative analysis ofdenitrifying activities of Hyphomicrobium nitrativorans, Hyphomicrobiumdenitrificans, and Hyphomicrobium zavarzinii. Appl Environ Microbiol81:5003–5014. doi:10.1128/AEM.00848-15.

Editor: V. Müller

Address correspondence to Richard Villemur, richard.villemur@iaf.inrs.ca.

* Present address: Christine Martineau, Laboratoire de Santé Publique du Québec,Ste-Anne-de-Bellevue, QC, Canada.

Supplemental material for this article may be found at http://dx.doi.org/10.1128/AEM.00848-15.

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AEM.00848-15

August 2015 Volume 81 Number 15 aem.asm.org 5003Applied and Environmental Microbiology

crobium species, H. nitrativorans, H. denitrificans, and H. zavarzi-nii, were compared at the genetic and physiological levels. BecauseH. nitrativorans was isolated from a marine environment, we fo-cused on the effect of salt on denitrification. Among the majordifferences found between their genomes was the type of nitratereductase that they encode. Our results showed that H. nitrativ-orans sustained better denitrifying activity than the other two spe-cies and had a better tolerance to higher nitrate and NaCl concen-trations. The effect of variations in the nitrate and NaClconcentrations on the relative expression levels of the nitrate andnitrite reductase genes was also assessed.

MATERIALS AND METHODSGenome sequence analyses. Hyphomicrobium nitrativorans strain NL23T

(ATCC BAA-2476), Hyphomicrobium denitrificans strain XT (ATCC51888), and Hyphomicrobium zavarzinii strain ZV622T (ATCC 27496)were used in this study. The general characteristics of the three bacterialspecies are presented in Table S1 in the supplemental material. Publiclyavailable genome sequences of H. nitrativorans NL23 (GenBank accessionnumber NC_022997) (22), H. denitrificans ATCC 51888 (GenBank acces-sion number NC_014313) (25), and H. zavarzinii ATCC 27496 (GenBankaccession numbers KB911255 to KB911269 for 15 scaffolds) were re-trieved from GenBank, imported into RAST (Rapid Annotation UsingSubsystem Technology) (26) for annotation, and processed in SEEDviewer (27) to identify denitrification gene clusters and associated regula-tory genes. Orthologous genes were determined with the “sequence-basedcomparison” tool in SEED viewer and refined by hand. Briefly, the de-duced amino acid sequences for all the annotated genes of one genomewere compared to those of the other two genomes by BLAST (automatedprocess in SEED viewer). This comparison was performed three timeswith a different genome set as the reference each time. Only genes thatresulted from reciprocal analysis were considered orthologous. The threegenomes were also annotated by the Integrated Microbial Genomes(IMG) comparative analysis system of the Joint Genome Institute (http://img.jgi.doe.gov/) (28). The majority of the differences in annotationsbetween RAST and IMG were in genes encoding hypothetical proteinswith unknown functions. However, both annotation systems providedsimilar results for orthologous genes. The results were derived from RASTbecause supplemental options were not available in IMG. A genome com-parison of the three species was further performed with MAUVE (29). Theaverage nucleotide identity (ANI) was determined for each pair of species(http://enve-omics.ce.gatech.edu/ani/). Deduced amino acid sequencesderived from the napA, narG, nirK, cnorB, and nosZ genes that were re-trieved from the genomes of the three Hyphomicrobium species, as well assequences of their closest relatives and representative sequences of theProteobacteria, were aligned by using ClustalW, and phylogenetic treeswere constructed by using the neighbor-joining algorithm (30) with thePoisson correction method in MEGA version 6.06 (31).

Bacterial strains and growth conditions. Bacteria were cultured inmodified Hyphomicrobium medium 337a (32) [composition (per liter),1.3 g KH2PO4, 1.13 g Na2HPO4, 0.5 g (NH4)2SO4, 0.2 g MgSO4·7H2O,3.09 mg CaCl2·2H2O, 2.0 mg FeSO4·7H2O, 1.0 mg Na2MoO4·2H2O, 0.88mg MnSO4·4H2O, and 10 �g CuCl2·2H2O (pH 7.5)] at 30°C. The NaClconcentrations were variable depending on the assay and are specifiedbelow. When required, nitrate was added to the growth medium as so-dium nitrate (NaNO3) at a concentration of 14.3 mM (200 mg NO3-Nliter�1, unless otherwise specified). Nitrate was added to the preculturesfor all of the assays conducted under denitrifying conditions (unless oth-erwise specified). Methanol was added to the growth medium as a carbonsource at a concentration of 0.3% (vol/vol) prior to inoculation. VitaminB12 was also added at a concentration of 0.1 mg liter�1 prior to inoculationto stimulate bacterial growth on solid medium and in liquid medium tostart cultures from bacterial colonies. The inoculum was made of freshaerobic liquid cultures. Bacterial cells were pelleted, dispersed in a 0.85%

NaCl solution, and homogenized prior to inoculation. The initial opticaldensity of the cultures was adjusted to reach an optical density at 600 nm(OD600) close to 0.1. In order to assess the capacity of the three Hyphomi-crobium species to achieve complete denitrification, the N2O concentra-tions in the headspace of cultures incubated with and without 10% (vol/vol) acetylene, an inhibitor of the nitrous oxide reductase, were compared(33).

Aerobic cultures used for RNA extraction were incubated with con-stant shaking (250 rpm) in 200-ml Erlenmeyer flasks containing 30 ml ofgrowth medium. For cultures under denitrifying conditions, 30 ml ofgrowth medium supplemented with nitrate was dispensed into 70-mlserum vials. The vials were sealed with rubber stoppers and metal seals,flushed for 10 min with N2 to remove oxygen from the headspace, andautoclaved. Substrate addition, inoculation, and sampling of these vialswere performed through the rubber stopper using a syringe and a needle.All cultures were incubated in the dark. Bacterial growth was monitoredby spectrophotometry (OD600). The nitrate and nitrite concentrations inthe cultures were determined by ion chromatography using an 850Professional IC instrument (Metrohm, Herisau, Switzerland) with aMetrosep A Supp 5 analytical column (250 by 4.0 mm), while N2O wasanalyzed by injecting 250 �l of the headspace into an Agilent 6890gas chromatograph equipped with a microelectron capture detector.

RNA extraction. RNA was extracted from 15 to 30 ml of bacterialculture in the mid-exponential growth phase. Bacterial cells were centri-fuged, transferred into 2-ml tubes containing 500 mg of 0.1-mm zirconia-silica beads (Biospec Products, Bartlesville, OK, USA), and dispersed in700 �l of a solution containing precooled TPM buffer (50 mM Tris-HCl[pH 7.0], 1.7% [wt/vol] polyvinylpyrrolidone K25, 20 mM MgCl2), 35 �lof 20% SDS, and 500 �l of phenol. The samples were flash frozen withliquid nitrogen and stored at �80°C until RNA extraction was performed.After thawing, RNA was extracted by bead beating twice with a Fast-Prep-24 instrument (MP Biomedicals, Solon, OH, USA), with the powerset at 6.5 for 45 s. The tubes were then centrifuged at 20,000 � g for 5 minat 4°C, and the upper phase was transferred into a new tube. The beadbeating and centrifugation steps were repeated following the addition of700 �l of PBB buffer (5 mM Tris-HCl [pH 7.0], 5 mM Na2 EDTA, 1%[wt/vol] SDS, 6% [vol/vol] water-saturated phenol) to the tubes contain-ing the bacterial cells. Both upper phases were further extracted with phe-nol-chloroform-isoamyl alcohol (25:24:1 [vol/vol/vol]) and chloroform-isoamyl alcohol (24:1 [vol/vol]). For each extraction step, centrifugationwas performed at 20,000 � g for 5 min at 4°C. RNA was recovered byprecipitation with a 1/10 volume of 3 M sodium acetate (pH 5.2) and 2volumes of 100% ethanol at �20°C overnight. After centrifugation at20,000 � g for 45 min at 4°C, the supernatant was discarded, and thepellets were washed with 70% ethanol (13,000 � g for 10 min at 4°C),dried at room temperature for 15 min, dissolved in diethyl pyrocarbonate(DEPC)-treated water, and treated with Turbo DNase (Ambion; LifeTechnologies Inc., Burlington, ON, Canada) for 30 min to remove allcontaminating DNA. Finally, RNA extracts were purified by using theRNeasy MinElute Cleanup kit according to the manufacturer’s instruc-tions (Qiagen, Toronto, ON, Canada). RNA integrity was verified by gelelectrophoresis, and RNA quantification was performed by using a Nano-Drop ND-1000 spectrophotometer (NanoDrop Products, Wilmington,DE, USA). The absence of DNA was confirmed by endpoint PCR analysisusing 16S rRNA gene universal primers.

Reverse transcription and quantitative PCR assays. cDNAs weregenerated from RNA by using hexameric primers and the reverse tran-scription system from Promega (Madison, WI, USA) with 1 �g of ex-tracted RNA, according to the manufacturer’s protocols. Real-time quan-titative PCR (qPCR) assays were performed with PerfeCTa SYBR greenSuperMix ROX (Quanta; VWR International, Ville Mont-Royal, QC,Canada). The final volume of all of the reaction mixtures was 20 �l, andthe amplification mix was composed of 10 �l of PerfeCTa SYBR greenSuperMix, 0.4 �l of each primer (40 pmol), 4.2 �l of RNA-free water, and5 �l of cDNA (50 ng). All of the reactions were performed with a Corbett

Martineau et al.

5004 aem.asm.org August 2015 Volume 81 Number 15Applied and Environmental Microbiology

Rotor-Gene 6000 real-time PCR machine (Qiagen). PCR started with aninitial denaturation step for 10 min at 95°C followed by 40 cycles of 30 s at95°C, 30 s at 55°C, and 30 s at 72°C. The fluorescence signal was acquiredat the end of the 72°C step and at the end of a 15-s reading step at 80°C thatwas added to the end of each cycle to avoid fluorescence from primerdimers. Primers were designed with Primer3web version 4.0.0 (34, 35) totarget the napA, narG, nirK, cnorB, and nosZ denitrification genes and therpoB and dnaG reference genes of the three Hyphomicrobium species (seeTable S2 in the supplemental material). Each of the primer sets was testedfor specificity and efficiency of amplification by qPCR with serial dilutionsof purified PCR fragments (efficiencies of between 0.92 and 1.05).

Calculations and statistical analyses. Growth and denitrificationrates were obtained by calculating the slope of the linear portion of thegrowth and denitrification curves, respectively. Relative gene expressionwas calculated according to the ��CT method (36). For each calculation,only one housekeeping gene was used as a control. For each Hyphomicro-bium species, the control gene was chosen from among the two genestested (rpoB and dnaG) by using RefFinder (http://www.leonxie.com/referencegene.php), an online tool that allows the selection of the appro-priate control gene based on the lowest variability according to geNorm(37), Normfinder (38), BestKeeper (39), and the comparative ��CT

method (40). Statistical analyses were performed with R (version 2.9.0; RFoundation for Statistical Computing). The effects of nitrate and NaClconcentrations on growth and denitrification rates were tested by usingfactorial analysis of variance (ANOVA). Prior to factorial ANOVA, nor-mality was tested by using the “shapiro.test” function. When necessary,data were log transformed to meet parametric ANOVA assumptions.ANOVA and post hoc Tukey honestly significant difference (HSD) testswere then carried out by using the “aov” and “TukeyHSD” functions,respectively. If significant interactions were detected between the two fac-tors, ANOVA was performed on a subset of data that were generated foreach factor. Normality was tested again on each data subset prior toANOVA. The statistical significance of the ��CT values was tested byStudent’s t test (two tailed, heteroscedastic).

RESULTSGenome comparison. The general characteristics of the three ge-nomes are presented in Tables 1 and 2 and Fig. 1. A total of 1,961orthologous genes are shared among the three species (Fig. 1),with 2,150 orthologous genes shared between H. nitrativorans andH. denitrificans, 2,687 shared between H. nitrativorans and H. za-varzinii, and 2,355 shared between H. denitrificans and H. zavar-zinii (Table 2). The number of unique genes (found in one speciesbut not in the other two) constitutes between 16.3 and 25.7% ofthe total genes, with the highest number of unique genes beingfound in H. denitrificans (Fig. 1). However, a large number(�75%) of these unique genes are annotated as hypothetical pro-teins with unknown functions (HP). When the genomes werecompared for the average deduced amino acid sequence identitybetween orthologous genes and for the average nucleotide identity(ANI) between whole-genome sequences, H. nitrativorans and H.zavarzinii showed a higher level of identity with each other thanwith H. denitrificans (Table 2). These results are consistent withthe results of phylogenetic analysis of the 16S rRNA gene se-quences (21).

Putative unique pathways were found in each species or in twoout of three species (see Table S3 in the supplemental material). Along cluster of genes associated with photosynthesis is present inH. zavarzinii, although there is no indication in the literature ofphotosynthesis-related activities in this bacterial species. The ureadecomposition pathways differ among the three species. Urea isdegraded into CO2 and NH3 by urease in H. denitrificans and byurea carboxylase and allophanate hydrolase in H. zavarzinii. Nourea decomposition pathway was found in H. nitrativorans.Genes encoding the Na�/H� multisubunit antiporter were foundin H. nitrativorans, and NhaA-type and NhaD-type Na�/H� an-tiporter genes were found in the other 2 species. H. nitrativoransand H. zavarzinii carry genes encoding methylamine dehydroge-nase and associated proteins. These genes were not found in H.denitrificans, which is surprising because this bacterium can utilizemonomethylamine, dimethylamine, and trimethylamine as solecarbon sources (5). Clustered regularly interspaced short palin-dromic repeats (CRISPRs) and CRISPR-associated proteins werefound in H. nitrativorans, whereas only CRISPR-associated pro-teins were found in H. zavarzinii. However, the draft genomeshowed an �3,000-bp gap flanking the chromosomal region ofthe genome, suggesting a repeated region that could not be assem-bled properly.

Genes associated with the denitrification pathway. All threeHyphomicrobium species studied here possess gene clusters en-coding the full set of enzymes necessary to achieve complete deni-trification, although they differ greatly in their nitrate reductases(Table 3). The genome of H. nitrativorans contains a gene clusterencoding the periplasmic nitrate reductase Nap, the genome of H.denitrificans contains a gene cluster encoding the membrane-

TABLE 1 General characteristics of the genomes of the threeHyphomicrobium speciesa

Characteristic

Value for species

H. nitrativorans H. zavarzinii H. denitrificans

Size (bp) 3,653,837 4,651,795 3,638,969GC content (%) 63.9 63.7 60.8No. of scaffolds 1 15 1No. of genes 3,599 4,399 3,702

No. of CDSs 3,548 4,348 3,653rRNA operons 3 3 3tRNA 48 48 46

No. of CDSs withpredicted function

2,182 2,764 2,293

No. of CDSs withoutpredicted function

1,366 1,584 1,360

a Sequence analyses were performed in RAST. CDSs, coding DNA sequences.

TABLE 2 Comparative analysis of the genomes of the Hyphomicrobium species

Species

H. nitrativorans H. zavarzinii

No. of orthologsa Ortholog identity (%)b ANI (%) No. of orthologsa Ortholog identity (%)b ANI (%)

H. zavarzinii 2,687 73.90 82.50H. denitrificans 2,150 60.60 80.80 2,355 61.80 81.00a Number of orthologous genes in common between the respective species.b Average percent deduced amino acid sequence identity between orthologous genes.

Denitrification in Three Hyphomicrobium Species

August 2015 Volume 81 Number 15 aem.asm.org 5005Applied and Environmental Microbiology

bound nitrate reductase Nar, and the genome of H. zavarziniicontains gene clusters encoding one Nap and two Nar enzymes.The three species also have a nirK gene encoding the copper-con-taining nitrite reductase, along with the nirV gene encoding anaccessory protein in H. nitrativorans and H. zavarzinii. Gene clus-ters encoding the nitric oxide (Nor) and nitrous oxide (Nos) re-ductases and with similar gene organizations were also detected inthe three genomes. The only exception is norE, which is locatedupstream in a direction opposite that of cnorCBQD in H. denitri-ficans (Table 3). Genes associated with nitrate transport were de-

tected upstream of the nar gene clusters in H. denitrificans and H.zavarzinii. Two distinct genes encoding NarK transporters, aNarK1-type symporter and a NarK2-type nitrate/nitrite anti-porter, are located upstream of the nar-2 gene cluster in H. zavar-zinii. One gene encoding a fused NarK1-NarK2-type transporter(here named narK12f) (41) is located upstream of the nar-1 genecluster in H. zavarzinii and the nar gene cluster in H. denitrificans.

Genes commonly associated with the regulation of denitrifica-tion gene expression, such as nnrS and nrrR, were detected in thethree genomes. Most of these genes belong to the CRP (cyclicAMP receptor protein)/FNR (fumarate and nitrate reductase reg-ulation) family; some of them are located close to the denitrifica-tion gene clusters (Table 3). The narXL two-component tran-scription regulatory system was found in H. denitrificans and H.zavarzinii. However, it is not located in the vicinity of the denitri-fication gene clusters in either species.

Phylogenetic analyses were performed on the deduced aminoacid sequences of genes encoding the catalytic subunits of thedifferent reductases (NapA, NarG, NirK, cNorB, and NosZ)found in the three Hyphomicrobium species. All these reductasesare phylogenetically related to known denitrification reductasesfound in other Alphaproteobacteria, with sequence identities gen-erally above 80%, except for NarG2 and NosZ of H. zavarzinii(maximum of 69% identity in both cases) (see Fig. S1 in the sup-plemental material).

Effect of nitrate and NaCl on bacterial growth and denitrifi-cation activity. The three Hyphomicrobium species cultivated un-der aerobic conditions with or without the addition of 14.3 mMNaNO3 showed similar growth curves. No nitrate reduction ornitrite production was detected under these conditions (data notshown). When incubated under denitrifying conditions at thesame nitrate concentration, the three Hyphomicrobium specieshad the capacity to grow and to completely eliminate nitrate with-out nitrite accumulation, although growth was much slower thanthat under aerobic conditions (Fig. 2). The addition of acetyleneled to the accumulation of N2O in the headspace of the vials, whileno (H. nitrativorans) or less (H. denitrificans and H. zavarzinii)N2O accumulation was observed in cultures incubated without

FIG 1 Venn diagram illustrating the number of orthologs found across thethree species. Numbers in the nonsharing zones represent unique genes withno homology to sequences in the other two species. Percentages of uniquegenes that were identified as encoding a HP (hypothetical protein with un-known function) are illustrated. In the shared zones, only genes that resultedfrom reciprocal analysis (e.g., H. nitrativorans versus H. denitrificans and H.denitrificans versus H. nitrativorans) were considered.

TABLE 3 Gene clusters associated with the dissimilatory pathway of denitrification

Reductase

Associated gene cluster(s) (locus tags)

H. nitrativorans (GenBank accession no.CP006912)

H. zavarzinii (GenBank accession no.NZ_ARTG01000000)

H. denitrificans (GenBank accessionno. CP002083)

Nar NP narK12f-narGHJI (nar-1)(F812RS0121405 to F812RS0121425)

narK12f-narGHJI (Hden0925 toHden0929)

narK1-narK2-narGHJI (nar-2)(F812RS24645, F812RS24650,F812RS0118690 to F812RS0118710)

Nap napEFDABC-napGH (W91113865 to W91113890,W91102520 to W91102525)

napDABC-napGH (F812RS0114960 toF812RS0114975, F812RS0108875 toF812RS0108880)

NPa

Nir nirK-nirV-nnrR (W91113050 to W91113060) nirK-nirV-nnrR (F812RS0115595 toF812RS0115605)

nirK-nnrS (Hden0590 andHden0591)

Nor nnrS-cnorCBQDE (W91107415 to W91107440) cnorCBQDE (F812RS0108785 toF812RS0108805)

cnorE-cnorCBQD (Hden0580 toHden0584)

Nos nosRZDFYLX (W91108030 to W91108060) nnrS-nosRZDFYLX (F812RS24275 toF812RS24285, F812RS0116365 toF812RS0116390)

nosRZDFYLX (Hden1877 toHden1883)

a NP, not present.

Martineau et al.

5006 aem.asm.org August 2015 Volume 81 Number 15Applied and Environmental Microbiology

acetylene (Fig. 3), indicating that the three species can achievecomplete denitrification.

The effect of increasing nitrate concentrations on the growthand denitrification kinetics of the three Hyphomicrobium specieswas assessed under denitrifying conditions, with the NaNO3 con-centration ranging from 14.3 to 107 mM (200 to 1,500 mg NO3-N/liter) (Fig. 2). H. nitrativorans had the capacity to grow (Fig. 2A)and denitrify without significant nitrite accumulation (Fig. 2D) atall of the concentrations of NaNO3 tested, with optimal growth at71.4 mM NaNO3. In contrast, growth and denitrification by H.

denitrificans and H. zavarzinii were almost completely inhibitedat 107 mM NaNO3 (Fig. 2B, C, E, and F). For H. denitrificans andH. zavarzinii, optimal growth and denitrification were observed at42.8 mM NaNO3, although higher levels of nitrite were con-stantly detected in cultures of H. denitrificans. While these higherlevels of nitrite were transient at initial concentrations of 14.3 and42.8 mM NaNO3 in H. denitrificans, nitrite accumulation leadingto incomplete denitrification was detected at 71.4 mM NaNO3

(Fig. 2E).Because H. nitrativorans was isolated from a marine environ-

FIG 2 Effect of nitrate concentration on growth and denitrification. H. nitrativorans (A and D), H. denitrificans (B and E), and H. zavarzinii (C and F) werecultured in triplicate under denitrifying conditions with an initial concentration of 14.3 mM (diamonds), 42.8 mM (squares), 71.4 mM (triangles), or 107 mM(crosses) NaNO3. At regular intervals, growth (A to C) and nitrate and nitrite concentrations (D to F) were measured. In panels D to F, full lines indicate nitrateconcentrations, and dotted lines indicate nitrite (NOx) concentrations. Assays were performed with an NaCl concentration of 0.5% in the growth medium. Thestandard deviations of the means for each time point are presented.

Denitrification in Three Hyphomicrobium Species

August 2015 Volume 81 Number 15 aem.asm.org 5007Applied and Environmental Microbiology

ment, tolerance to increasing NaCl concentrations ranging from 0to 2% NaCl was assessed under denitrifying conditions for thethree species. Growth of and denitrification by H. nitrativoranswere detected at all of the NaCl concentrations tested, withoutnitrite accumulation, although a concentration of 2% led toslower growth and denitrification (Fig. 4A and D). IncreasingNaCl concentrations ranging from 0 to 1% led to delayed growthand denitrification in H. denitrificans (Fig. 4B and E) and slowergrowth and denitrification in H. zavarzinii (Fig. 4C and F). Atransient accumulation of nitrite was also observed in H. denitri-ficans cultures. At 2% NaCl, these two species showed poor or nogrowth and no denitrifying activity.

To further evaluate whether the interaction between nitrateand NaCl could lead to significant differences in growth anddenitrification rates in the three Hyphomicrobium species, wecarried out a factorial experiment using two different concentra-tions of each of the two factors in question. Growth and denitri-fication were assessed at 14.3 and 71.4 mM NaNO3 and at 0 and0.5% NaCl (Table 4). The results showed that the three speciesresponded differently to variations of nitrate and NaCl levels.While no significant interactions between the nitrate and NaClconcentrations were detected for H. nitrativorans and H. denitri-ficans, significant interactions between these two factors were de-tected for the growth rate and the denitrification rate in H. zavar-zinii (Table 4). Single-factor analysis of the data subsets showedthat growth and denitrification rates in H. zavarzinii were nega-tively affected by higher nitrate and NaCl concentrations, with thelowest rates being detected at 71.4 mM NaNO3 and 0.5% NaCl(Tables 5 and 6). In H. nitrativorans, while nitrate and NaCl hadno significant effect on the growth rate, a significant effect of bothfactors on the denitrification rate was detected, with higher valuesbeing observed at higher nitrate and NaCl concentrations (Tables5 and 6). In contrast, neither of the two factors had a significanteffect on the denitrification rate in H. denitrificans (Table 6), butthe nitrate concentration had a significant effect on the growthrate, with a considerably lower value at 71.4 mM NaNO3 than at14.3 mM NaNO3 (Table 5).

Effect of nitrate, NaCl, and oxygen on relative gene expres-sion levels of the nitrate and nitrite reductase genes. Our resultsshowed the influence of nitrate and NaCl on the growth and deni-trification of the three Hyphomicrobium species. The effects ofthese parameters were also assessed at the nitrate and nitrite re-ductase gene expression levels. We focused our analysis on thesegenes because the three species differ greatly in their nitrate reduc-tases; H. denitrificans also harbors a very distinct nitrite reductase,while NO and N2O reductases found in the three species are moresimilar.

The three Hyphomicrobium species were cultured under thefollowing denitrifying conditions: (i) 14.3 mM NaNO3 and 0%NaCl, (ii) 71.4 mM NaNO3 and 0% NaCl, (iii) 14.3 mM NaNO3

and 0.5% NaCl, and (iv) 71.4 mM NaNO3 and 0.5% NaCl. Rela-tive expression levels of the napA, narG, and nirK genes were mea-sured during the mid-exponential growth phase, with culturesgrown in 14.3 mM NaNO3 and 0% NaCl as the reference culture(Fig. 5A to C). In H. nitrativorans, higher nitrate and NaCl con-centrations led to a small (1.4- to 3.4-fold) but significant increasein the relative gene expression levels of napA and nirK (Fig. 5A). InH. denitrificans, a significant effect was observed exclusively forrelative gene expression levels of nirK in cultures grown with 71.4mM NaNO3 and 0% NaCl (5.8-fold increase) (Fig. 5B). In H.zavarzinii, the nitrate and NaCl concentrations affected the threenitrate reductase genes to different degrees. The relative gene ex-pression level of napA did not change under any of the conditionstested. Both factors had a negative effect on narG1 gene expressionbut a positive effect on narG2 gene expression (Fig. 5C). Similar tonarG1, the relative gene expression level of nirK in H. zavarziniiwas negatively affected by higher nitrate and NaCl concentrations.

Oxygen is also known to have a major effect on denitrificationand the denitrification reductases. The three Hyphomicrobiumspecies were cultured under aerobic conditions, under aerobicconditions with 14.3 mM NaNO3, or under denitrifying condi-tions with 14.3 mM NaNO3. The relative expression levels of thenapA, narG, and nirK genes were measured as described above,with the culture grown aerobic conditions set as the referenceculture (Fig. 5D and E). In H. nitrativorans, the addition of nitrateto aerobic cultures led to small but significant increases in therelative gene expression levels of napA and nirK (1.8- and 3.2-fold,respectively), while incubation under denitrifying conditions ledto much higher relative gene expression levels for both genes (7.1-and 35.1-fold, respectively) (Fig. 5D). In H. denitrificans, the onlydifference observed was for the relative gene expression level ofnarG, which was significantly higher in the presence of nitrateunder aerobic conditions (Fig. 5E). In H. zavarzinii, none of thegenes tested was significantly affected by the addition of nitrate toaerobic cultures (Fig. 5F). However, the incubation of this bacte-rial species under denitrifying conditions led to a strong increasein the relative gene expression level of narG1 (36.4-fold), a slightbut significant increase in the level of narG2 (1.8-fold), and asignificant decrease in the level of napA (�8.3-fold) (Fig. 5F).

DISCUSSION

H. nitrativorans, H. denitrificans, and H. zavarzinii possess genesencoding the full set of enzymes required to perform completedenitrification, and they were able to achieve this process. Underthe culture conditions tested, H. nitrativorans was found to be themost efficient of the three Hyphomicrobium species evaluated.Compared to the two other species, it has the capacity to denitrify

FIG 3 Effect of acetylene on the production of N2O. H. nitrativorans (tri-angles), H. denitrificans (squares), and H. zavarzinii (circles) were cultured intriplicate under denitrifying conditions with (full lines) or without (dottedlines) acetylene at an initial concentration of 14.3 mM NaNO3. At regularintervals, the nitrous oxide concentration in the headspace was measured.

Martineau et al.

5008 aem.asm.org August 2015 Volume 81 Number 15Applied and Environmental Microbiology

at high nitrate concentrations, without nitrite accumulation.Consistent with this observation, an increase in the nitrate con-centration in the growth medium induced significant increases innarG and nirK relative gene expression levels in H. nitrativorans.In contrast, an increase of the initial nitrate concentration in thegrowth medium led to nitrite accumulation in H. denitrificanscultures despite a significant increase in the nirK relative gene

expression level at a higher nitrate concentration. Nitrate inhibi-tion of nitrite reduction was previously reported for several bac-teria (42–44) and has been associated with a phenomenon of com-petition between nitrate and nitrite reductases for electrondonors. These results suggest that in H. denitrificans, the nitratereductase might have a higher affinity for electrons than the nitritereductase. H. denitrificans is characterized by its unique nitrite re-

FIG 4 Effect of NaCl concentration on growth and denitrification. H. nitrativorans (A and D), H. denitrificans (B and E), and H. zavarzinii (C and F) werecultured in triplicate under denitrifying conditions with 0% (diamonds), 0.5% (squares), 1% (triangles), and 2% (crosses) NaCl. At regular intervals, growth (Ato C) and nitrate and nitrite (NOx) concentrations (D to F) were measured. In panels D to F, full lines indicate nitrate concentrations, and dotted lines indicatenitrite concentrations. Standard deviations of the means for each time point are presented.

Denitrification in Three Hyphomicrobium Species

August 2015 Volume 81 Number 15 aem.asm.org 5009Applied and Environmental Microbiology

ductase, which is clearly distinct from the nitrite reductases foundin other Hyphomicrobium species. It is characterized by the pres-ence of two type I Cu sites, as opposed to the single type I Cu sitefound in the nitrite reductases of most bacteria (16, 17, 45). Therole of this second type I Cu site has not been elucidated yet.Currently, the role of this distinct nitrite reductase in the denitri-fying capacity of H. denitrificans is unknown, but it might belinked to the nitrite accumulation observed in H. denitrificans cul-tures. Growth of H. denitrificans and H. zavarzinii was totally in-hibited at the highest nitrate concentration tested. Because nitratewas supplied as NaNO3, it is possible that this growth inhibitionwas exacerbated by the input of sodium in the growth medium, aspreviously observed for Shewanella oneidensis cultures (46).

As expected, H. nitrativorans also demonstrated better toler-ance to NaCl than the two other Hyphomicrobium species testedand showed small but significant increases in the napA and nirKrelative gene expression levels when NaCl was added to the growthmedium. More surprisingly, H. nitrativorans showed better toler-ance to NaCl when cultured under denitrifying conditions thanpreviously reported for aerobic growth. Indeed, while total growthinhibition at NaCl concentrations of �1% was previously re-ported for H. nitrativorans under aerobic conditions (21), ourresults demonstrate growth and complete denitrification by thisbacterium at NaCl concentrations of up to 2%. The mechanismsof this enhanced tolerance to NaCl could be associated with themultisubunit Na�/H� antiporter. A gene cluster encoding thistype of antiporter was found in the H. nitrativorans genome,whereas genes encoding an NhaA-type Na�/H� antiporter werefound in the genomes of the two other species. The multisubunitNa�/H� antiporter has been associated with high Na� extrusionactivity in halotolerant Staphylococcus aureus (47). Despite havingbetter NaCl tolerance than the other Hyphomicrobium species, H.nitrativorans remains quite sensitive to NaCl compared to otherbacteria isolated from denitrifying biofilms, such as Methyl-

ophaga nitratireducenticrescens JAM1, which can tolerate NaClconcentrations of up to 8% (24). We hypothesize that H. nitrativ-orans benefits from the osmoprotection of other bacteria, such asM. nitratireducenticrescens JAM1, to thrive in the artificial seawa-ter of the Biodome aquarium (NaCl concentration of 2.4%). Thishypothesis is currently being studied in our laboratory.

The detailed comparison of the H. nitrativorans, H. denitrifi-cans, and H. zavarzinii genomes has highlighted the facts that alarge number of orthologous genes are shared among these threespecies and that very few genes with known functions are uniqueto a single species. From this perspective, the fact that the threespecies differ greatly in their nitrate reductases is noteworthy, andit can be hypothesized that the differences observed in their deni-trification capacities reflect these variations observed at the ge-netic level. Indeed, while H. nitrativorans and H. denitrificans arecharacterized by the presence of a single Nap or Nar gene cluster,respectively, one Nap and two Nar gene clusters have been iden-tified in the genome of H. zavarzinii. Such an occurrence of mul-tiple operons encoding nitrate reductases has previously beenreported for other microorganisms. One well-characterized ex-ample is E. coli K-12, which also expresses two different cytoplas-mic membrane nitrate reductases (NRA and NRZ), encoded bynarGHJI and narZYWV, respectively, and one periplasmic nitratereductase, encoded by napFDAGHBC (48–50). Gene deletions inE. coli K-12 indicate that Nap can support anaerobic growth whenit is the only nitrate reductase available, although NRA is clearlythe main enzyme responsible for nitrate reduction when all en-zymes are present (48).

While Nap in E. coli is expressed under anaerobic conditions(50) and is specifically involved in nitrate reduction at low nitrateconcentrations (48, 51), different functions have been proposedfor the Nap system in other bacteria. In Paracoccus spp., in whichboth the Nar and Nap systems are present, Nar is expressed underanaerobic conditions, while Nap is expressed during aerobic

TABLE 4 Factorial ANOVA of the effects of nitrate and NaCl concentrations on growth and denitrification ratesa

Source of variation

P value

H. nitrativorans H. denitrificans H. zavarzinii

Growth rateDenitrificationrate Growth rate

Denitrificationrate Growth rate

Denitrificationrate

Nitrate NS 0.0041 0.0042 NS 0.001 0.001NaCl NS 0.0178 NS NS 0.001 0.001Nitrate � NaCl NS NS NS NS 0.0145 0.0087a The nitrate concentrations tested were 14.3 and 71.4 mM NaNO3; the NaCl concentrations tested were 0 and 0.5%. P values of 0.05 are considered statistically significant. NS,not significant.

TABLE 5 Effect of nitrate and NaCl concentrations on growth rates of three Hyphomicrobium speciesa

NaNO3 concn (mM)

Growth rate (OD units h�1)

H. nitrativorans H. denitrificans H. zavarzinii

0% NaCl 0.5% NaCl Mean 0% NaCl 0.5% NaCl Mean 0% NaCl 0.5% NaCl Mean

14.3 0.00206 0.00249 0.00228 0.00325 0.00277 0.00301a 0.00357Aa 0.00287Ba NA71.4 0.00220 0.00225 0.00223 0.00234 0.00228 0.00231b 0.00291Ab 0.00138Bb NA

Mean 0.00213 0.00237 0.00280 0.00253 NA NAa Uppercase letters indicate significant differences between columns according to Tukey’s test; lowercase letters indicate significant differences between rows according to Tukey’stest. NA, not applicable.

Martineau et al.

5010 aem.asm.org August 2015 Volume 81 Number 15Applied and Environmental Microbiology

growth, in the presence or absence of nitrate (13, 52, 53). Thisexpression pattern is similar to our results for H. zavarzinii, withgrowth under denitrifying conditions leading to the repression ofnapA and the overexpression of narG1 and, to a lesser extent,

narG2. In Paracoccus spp., Nap is responsible for aerobic nitratereduction, and it has been suggested that this activity might helpdissipate excess reductant, which is further supported by the factthat Nap is induced by growth on highly reduced carbon sub-

TABLE 6 Effect of nitrate and NaCl concentrations on denitrification rates of three Hyphomicrobium speciesa

NaNO3 concn (mM)

Denitrification rate (mM h�1)

H. nitrativorans H. denitrificans H. zavarzinii

0% NaCl 0.5% NaCl Mean 0% NaCl 0.5% NaCl Mean 0% NaCl 0.5% NaCl Mean

14.3 0.121 0.174 0.148b 0.126 0.114 0.120 0.280Aa 0.220Ba NA71.4 0.186 0.204 0.196a 0.117 0.140 0.129 0.166Ab 0.123Bb NA

Mean 0.154A 0.189A 0.122 0.127 NA NAa Uppercase letters indicate significant differences between columns according to Tukey’s test; lowercase letters indicate significant differences between rows according to Tukey’stest. NA, not applicable.

FIG 5 Relative expression levels of the nitrate and nitrite reductase genes in the three species cultured under different conditions. (A to C) Effects of nitrate andNaCl concentrations. The three species were cultured in triplicate under denitrifying conditions with 14.3 mM NaNO3 and 0% NaCl (reference cultures arerepresented by the dashed line set at 1 in each panel), with 71.4 mM NaNO3 and 0% NaCl, with 14.3 mM NaNO3 and 0.5% NaCl, or with 71.4 mM NaNO3 and0.5% NaCl. (D and E) Effects of oxygen. The three species were cultured in triplicate under aerobic conditions (reference cultures are represented by the dashedline set at 1 in each panel), aerobic conditions with 14.3 mM NaNO3, or denitrifying conditions with 14.3 mM NaNO3. Aerobic precultures for this assay wereprepared without adding nitrate to the growth medium. ***, P 0.001; **, 0.01 � P � 0.001; *, 0.05 � P � 0.01.

Denitrification in Three Hyphomicrobium Species

August 2015 Volume 81 Number 15 aem.asm.org 5011Applied and Environmental Microbiology

strates (13, 52). Despite higher levels of Nap expression underaerobic conditions than under denitrifying conditions, we did notdetect significant nitrate reduction in aerobic liquid cultures of H.zavarzinii. Similarly, the expression level of the Nap- operon inRhodobacter sphaeroides 2.4.3 was higher in aerobic liquid culturesthan in anaerobic cultures, without any detectable nitrate reduc-tion (54). However, aerobic reduction of nitrate into nitrite oc-curred when R. sphaeroides 2.4.3 was cultivated on solid medium,indicating a role of Nap- in redox homeostasis specific for theseculture conditions. Preliminary assays performed in our labora-tory indicated that this is not the case for H. zavarzinii, but moreextensive work will have to be performed to rule out this hypoth-esis.

While it is not uncommon to find multiple nitrate reductasesin a single bacterium, the presence of multiple nitrate reductasesof the same type has been reported only rarely in previous studies.Rhodobacter sphaeroides 2.4.3 possesses two Nap operons, one,identified as Nap-, being similar to Nap operons found in otherAlphaproteobacteria (napKEFDABC) and the other, identified asNap-�, being similar to the Nap operons found in Gammaproteo-bacteria (napFDAGHBC) such as Shewanella spp. and E. coli (54).These two distinct Nap isoforms have also been reported for sev-eral strains of Shewanella (55, 56). Recently, we also described thenew bacterial species Methylophaga nitratireducenticrescens JAM1(Gammaproteobacteria), which expresses two distinct Nar nitratereductases (Nar1 and Nar2) encoded by two narGHJI operons(23, 24, 57). For this bacterium, phylogenetic analyses revealedthat Nar1 is more closely related to the Nar system of Betaproteo-bacteria, whereas Nar2 is more closely related to the Nar systemfound in other Gammaproteobacteria, suggesting that the Nar1system was acquired by horizontal gene transfer (23). In H. zavar-zinii, while narG1 is closely related to narG sequences found inother Alphaproteobacteria, narG2 is only distantly related to anyknown narG sequence. The expression patterns of these two geneswere also slightly different, with a much stronger effect of anaer-obic conditions on the expression of narG1 than on the expressionof narG2. Further work, including the production of gene knock-out mutants for the different nitrate reductases in H. zavarzinii, isnecessary to completely elucidate the physiological roles played bythese enzymes.

Although the napA gene product in H. nitrativorans is morerelated to Nap- than to Nap-� of Rhodobacter sphaeroides 2.4.3,a distinct napA expression pattern was observed. In H. nitrativ-orans, napA gene expression levels were significantly higher in thepresence of nitrate and under denitrifying conditions than in theabsence of nitrate and under aerobic conditions. Similar resultshave been reported for Bradyrhizobium japonicum, which alsopossesses a single periplasmic nitrate reductase of the isoform(58). Although results from studies on Paracoccus pantotrophusand Rhodobacter sphaeroides point toward a specific role of theNap- isoform in aerobic denitrification and redox balancing,both the and � isoforms have been associated with anaerobicnitrate respiration in Shewanella spp. (55, 56). It is important tonote that the highest denitrification efficiencies were observed fora Shewanella strain lacking the nap-� operon (55) and for a nap-�mutant Shewanella strain (56), indicating that the isoform playsa major role in anaerobic respiration in this bacterial genus. Takentogether, these results further illustrate that there is no clear asso-ciation between the isoform and the physiological role of Nap and

that other factors, such as the presence of distinct enzymes withsimilar functions, might also play an important role.

Denitrifying conditions were shown to increase the relativegene expression level of nirK in H. nitrativorans. However, such aneffect was not observed for H. denitrificans and H. zavarzinii. Noprevious studies assessed quantitatively the expression of nirK un-der different culture conditions in Hyphomicrobium spp. Mi-croaerobic or anaerobic conditions are known to regulate nirKexpression in many bacterial species (59–61). In Rhodobactersphaeroides 2.4.3, nirK expression is regulated by NnrR, a tran-scriptional activator of the CRP/FNR family (60, 62). Gene anno-tation of the H. denitrificans genome did not reveal sequencesclosely related to nnrR, which could explain the absence of regu-lation of nirK expression. The lack of an effect of denitrifyingconditions on H. zavarzinii is surprising, as nnrR was founddownstream of nirK. In fact, H. zavarzinii and H. nitrativoranshave the same gene arrangement in the nirK locus (Table 3), andboth showed a potential NnrR binding site (data not shown). Fur-ther work will be required in order to elucidate this phenomenon.

ACKNOWLEDGMENTS

We thank Karla Vasquez for her technical assistance.This project was financially supported by a grant to R.V. from the

Natural Sciences and Engineering Research Council of Canada and byscholarships to C.M. from the Fondation Universitaire Armand-Frappierand the Fonds de Recherche du Québec Nature et Technologies.

REFERENCES1. Gliesche CG, Fesefeldt A, Hirsch P. 2005. Genus Hyphomicrobium

Stutzer and Hartleb 1898, 76AL, p 476 – 494. In Staley JT, Bryant MP,Pfennig N, Holt JG (ed), Bergey’s manual of systematic bacteriology. Wil-liams & Wilkins, Baltimore, MD.

2. Fesefeldt A, Kloos K, Bothe H, Lemmer H, Gliesche CG. 1998. Distri-bution of denitrification and nitrogen fixation genes in Hyphomicrobiumspp. and other budding bacteria. Can J Microbiol 44:181–186. http://dx.doi.org/10.1139/w97-139.

3. Kloos K, Fesefeldt A, Gliesche CG, Bothe H. 1995. DNA-probing indi-cates the occurrence of denitrification and nitrogen fixation genes in Hy-phomicrobium. Distribution of denitrifying and nitrogen fixing isolates ofHyphomicrobium in a sewage treatment plant. FEMS Microbiol Ecol 18:205–213. http://dx.doi.org/10.1111/j.1574-6941.1995.tb00177.x.

4. Timmermans P, Van Haute A. 1983. Denitrification with methanol:fundamental study of the growth and denitrification capacity of Hypho-microbium sp. Water Res 17:1249 –1255. http://dx.doi.org/10.1016/0043-1354(83)90249-X.

5. Urakami T, Sasaki J, Suzuki K-I, Komagata K. 1995. Characterizationand description of Hyphomicrobium denitrificans sp. nov. Int J Syst Bacte-riol 45:528 –532. http://dx.doi.org/10.1099/00207713-45-3-528.

6. Harder W, Attwood MM. 1978. Biology, physiology and biochemistry ofHyphomicrobia. Adv Microb Physiol 17:303–359. http://dx.doi.org/10.1016/S0065-2911(08)60060-0.

7. Baytshtok V, Lu H, Park H, Kim S, Yu R, Chandran K. 2009. Impact ofvarying electron donors on the molecular microbial ecology and biokinet-ics of methylotrophic denitrifying bacteria. Biotechnol Bioeng 102:1527–1536. http://dx.doi.org/10.1002/bit.22213.

8. Satoh H, Yamakawa T, Kindaichi T, Ito T, Okabe S. 2006. Communitystructures and activities of nitrifying and denitrifying bacteria in industrialwastewater-treating biofilms. Biotechnol Bioeng 94:762–772. http://dx.doi.org/10.1002/bit.20894.

9. Isaka K, Kimura Y, Osaka T, Tsuneda S. 2012. High-rate denitrificationusing polyethylene glycol gel carriers entrapping heterotrophic denitrify-ing bacteria. Water Res 46:4941– 4948. http://dx.doi.org/10.1016/j.watres.2012.05.050.

10. Baytshtok V, Kim S, Yu R, Park H, Chandran K. 2008. Molecular andbiokinetic characterization of methylotrophic denitrification using nitrateand nitrite as terminal electron acceptors. Water Sci Technol 58(2):359 –365. http://dx.doi.org/10.2166/wst.2008.391.

Martineau et al.

5012 aem.asm.org August 2015 Volume 81 Number 15Applied and Environmental Microbiology

11. Knowles R. 1982. Denitrification. Microbiol Rev 46:43–70.12. Philippot L. 2002. Denitrifying genes in bacterial and archaeal genomes.

Biochim Biophys Acta 1577:355–376. http://dx.doi.org/10.1016/S0167-4781(02)00420-7.

13. Richardson DJ, Berks BC, Russell DA, Spiro S, Taylor CJ. 2001. Func-tional, biochemical and genetic diversity of prokaryotic nitrate reductases.Cell Mol Life Sci 58:165–178. http://dx.doi.org/10.1007/PL00000845.

14. Kraft B, Strous M, Tegetmeyer HE. 2011. Microbial nitrate respiration—genes, enzymes and environmental distribution. J Biotechnol 155:104 –117. http://dx.doi.org/10.1016/j.jbiotec.2010.12.025.

15. Deligeer, Fukunaga R, Kataoka K, Yamaguchi K, Kobayashi K,Tagawa S, Suzuki S. 2002. Spectroscopic and functional characteriza-tion of Cu-containing nitrite reductase from Hyphomicrobium denitri-ficans A3151. J Inorg Biochem 91:132–138. http://dx.doi.org/10.1016/S0162-0134(02)00442-7.

16. Yamaguchi K, Kataoka K, Kobayashi M, Itoh K, Fukui A, Suzuki S.2004. Characterization of two type 1 Cu sites of Hyphomicrobium denitri-ficans nitrite reductase: a new class of copper-containing nitrite reducta-ses. Biochemistry 43:14180 –14188. http://dx.doi.org/10.1021/bi0492657.

17. Nojiri M, Xie Y, Inoue T, Yamamoto T, Matsumura H, Kataoka K,Deligeer, Yamaguchi K, Kai Y, Suzuki S. 2007. Structure and function ofa hexameric copper-containing nitrite reductase. Proc Natl Acad SciU S A 104:4315– 4320. http://dx.doi.org/10.1073/pnas.0609195104.

18. Yamaguchi K, Kawamura A, Ogawa H, Suzuki S. 2003. Characterizationof nitrous oxide reductase from a methylotrophic denitrifying bacterium,Hyphomicrobium denitrificans A3151. J Biochem 134:853– 858. http://dx.doi.org/10.1093/jb/mvg211.

19. Jérôme V, Hermann M, Hilbrig F, Freitag R. 2007. Development of afed-batch process for the production of a dye-linked formaldehyde dehy-drogenase in Hyphomicrobium zavarzinii ZV 580. Appl Microbiol Bio-technol 77:779 –788. http://dx.doi.org/10.1007/s00253-007-1218-z.

20. Klein CR, Kesseler FP, Perrei C, Frank J, Duine JA, Schwartz AC. 1994.A novel dye-linked formaldehyde dehydrogenase with some propertiesindicating the presence of a protein-bound redox-active quinone cofactor.Biochem J 301:289 –295.

21. Martineau C, Villeneuve C, Mauffrey F, Villemur R. 2013. Hyphomi-crobium nitrativorans sp. nov., isolated from the biofilm of a methanol-feddenitrification system treating seawater at the Montreal Biodome (Can-ada). Int J Syst Evol Microbiol 63:3777–3781. http://dx.doi.org/10.1099/ijs.0.048124-0.

22. Martineau C, Villeneuve C, Mauffrey F, Villemur R. 2014. Completegenome sequence of Hyphomicrobium nitrativorans strain NL23, a deni-trifying bacterium isolated from biofilm of a methanol-fed denitrificationsystem treating seawater at the Montreal Biodome. Genome Announc2(1):e01165-13 http://dx.doi.org/10.1128/genomeA.01165-13.

23. Auclair J, Lepine F, Parent S, Villemur R. 2010. Dissimilatory reductionof nitrate in seawater by a Methylophaga strain containing two highlydivergent narG sequences. ISME J 4:1302–1313. http://dx.doi.org/10.1038/ismej.2010.47.

24. Villeneuve C, Martineau C, Mauffrey F, Villemur R. 2013. Methyl-ophaga nitratireducenticrescens sp. nov. and Methylophaga frappieri sp.nov., isolated from the biofilm of the methanol-fed denitrification systemtreating the seawater at the Montreal Biodome. Int J Syst Evol Microbiol63:2216 –2222. http://dx.doi.org/10.1099/ijs.0.044545-0.

25. Brown PJB, Kysela DT, Buechlein A, Hemmerich C, Brun YV. 2011.Genome sequences of eight morphologically diverse alphaproteobacteria.J Bacteriol 193:4567– 4568. http://dx.doi.org/10.1128/JB.05453-11.

26. Aziz R, Bartels D, Best A, DeJongh M, Disz T, Edwards R, Formsma K,Gerdes S, Glass E, Kubal M, Meyer F, Olsen G, Olson R, Osterman A,Overbeek R, McNeil L, Paarmann D, Paczian T, Parrello B, Pusch G,Reich C, Stevens R, Vassieva O, Vonstein V, Wilke A, Zagnitko O. 2008.The RAST server: rapid annotations using subsystems technology. BMCGenomics 9:75. http://dx.doi.org/10.1186/1471-2164-9-75.

27. Overbeek R, Olson R, Pusch GD, Olsen GJ, Davis JJ, Disz T, EdwardsRA, Gerdes S, Parrello B, Shukla M, Vonstein V, Wattam AR, Xia F,Stevens R. 2014. The SEED and the Rapid Annotation of Microbial Ge-nomes Using Subsystems Technology (RAST). Nucleic Acids Res 42:D206 –D214. http://dx.doi.org/10.1093/nar/gkt1226.

28. Markowitz VM, Chen I-MA, Palaniappan K, Chu K, Szeto E, Pillay M,Ratner A, Huang J, Woyke T, Huntemann M, Anderson I, Billis K,Varghese N, Mavromatis K, Pati A, Ivanova NN, Kyrpides NC. 2014.IMG 4 version of the integrated microbial genomes comparative analysis

system. Nucleic Acids Res 42:D560 –D567. http://dx.doi.org/10.1093/nar/gkt963.

29. Darling AE, Mau B, Perna NT. 2010. progressiveMauve: multiple ge-nome alignment with gene gain, loss and rearrangement. PLoS One5:e11147 http://dx.doi.org/10.1371/journal.pone.0011147.

30. Saitou N, Nei M. 1987. The neighbor-joining method: a new method forreconstructing phylogenetic trees. Mol Biol Evol 4:406 – 425.

31. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. 2013. MEGA6:Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol 30:2725–2729. http://dx.doi.org/10.1093/molbev/mst197.

32. Atlas RM. 2010. Handbook of microbiological media, 4th ed, vol 1, p769 – 868. CRC Press, Boca Raton, FL.

33. Yoshinari T, Knowles R. 1976. Acetylene inhibition of nitrous oxidereduction by denitrifying bacteria. Biochem Biophys Res Commun 69:705–710. http://dx.doi.org/10.1016/0006-291X(76)90932-3.

34. Rozen S, Skaletsky H. 1999. Primer3 on the WWW for general usersand for biologist programmers, p 365–386. In Misener S, Krawetz S(ed), Bioinformatics methods and protocols, vol 132. Humana Press,Totowa, NJ.

35. Untergasser A, Nijveen H, Rao X, Bisseling T, Geurts R, LeunissenJAM. 2007. Primer3Plus, an enhanced Web interface to Primer3. NucleicAcids Res 35:W71–W74. http://dx.doi.org/10.1093/nar/gkm306.

36. Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression datausing real-time quantitative PCR and the 2-��Ct method. Methods 25:402– 408. http://dx.doi.org/10.1006/meth.2001.1262.

37. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De PaepeA, Speleman F. 2002. Accurate normalization of real-time quantitativeRT-PCR data by geometric averaging of multiple internal control genes.Genome Biol 3:RESEARCH0034. http://dx.doi.org/10.1186/gb-2002-3-7-research0034.

38. Andersen CL, Jensen JL, Orntoft TF. 2004. Normalization of real-timequantitative reverse transcription-PCR data: a model-based variance esti-mation approach to identify genes suited for normalization, applied tobladder and colon cancer data sets. Cancer Res 64:5245–5250. http://dx.doi.org/10.1158/0008-5472.CAN-04-0496.

39. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP. 2004. Determinationof stable housekeeping genes, differentially regulated target genes andsample integrity: BestKeeper—Excel-based tool using pair-wise correla-tions. Biotechnol Lett 26:509 –515. http://dx.doi.org/10.1023/B:BILE.0000019559.84305.47.

40. Silver N, Best S, Jiang J, Thein SL. 2006. Selection of housekeeping genesfor gene expression studies in human reticulocytes using real-time PCR.BMC Mol Biol 7:33. http://dx.doi.org/10.1186/1471-2199-7-33.

41. Goddard AD, Moir JWB, Richardson DJ, Ferguson SJ. 2008. Inter-dependence of two NarK domains in a fused nitrate/nitrite transporter.Mol Microbiol 70:667– 681. http://dx.doi.org/10.1111/j.1365-2958.2008.06436.x.

42. Polcyn W, Lucinski R. 2003. Aerobic and anaerobic nitrate and nitritereduction in free-living cells of Bradyrhizobium sp. (Lupinus). FEMSMicrobiol Lett 226:331–337. http://dx.doi.org/10.1016/S0378-1097(03)00620-7.

43. Almeida JS, Reis MAM, Carrondo MJT. 1995. Competition betweennitrate and nitrite reduction in denitrification by Pseudomonas fluorescens.Biotechnol Bioeng 46:476 – 484. http://dx.doi.org/10.1002/bit.260460512.

44. Kornaros M, Zafiri C, Lyberatos G. 1996. Kinetics of denitrification byPseudomonas denitrificans under growth conditions limited by carbonand/or nitrate or nitrite. Water Environ Res 68:934 –945. http://dx.doi.org/10.2175/106143096X127947.

45. Ellis MJ, Grossmann JG, Eady RR, Hasnain SS. 2007. Genomic analysisreveals widespread occurrence of new classes of copper nitrite reductases.J Biol Inorg Chem 12:1119 –1127. http://dx.doi.org/10.1007/s00775-007-0282-2.

46. Zhang H, Fu H, Wang J, Sun L, Jiang Y, Zhang L, Gao H. 2013. Impactsof nitrate and nitrite on physiology of Shewanella oneidensis. PLoS One8:e62629. http://dx.doi.org/10.1371/journal.pone.0062629.

47. Hiramatsu T, Kodama K, Kuroda T, Mizushima T, Tsuchiya T. 1998. Aputative multisubunit Na�/H� antiporter from Staphylococcus aureus. JBacteriol 180:6642– 6648.

48. Stewart V, Lu Y, Darwin AJ. 2002. Periplasmic nitrate reductase(NapABC enzyme) supports anaerobic respiration by Escherichia coliK-12. J Bacteriol 184:1314 –1323. http://dx.doi.org/10.1128/JB.184.5.1314-1323.2002.

49. Potter LC, Millington P, Griffiths L, Thomas GH, Cole JA. 1999.

Denitrification in Three Hyphomicrobium Species

August 2015 Volume 81 Number 15 aem.asm.org 5013Applied and Environmental Microbiology

Competition between Escherichia coli strains expressing either a periplas-mic or a membrane-bound nitrate reductase: does Nap confer a selectiveadvantage during nitrate-limited growth? Biochem J 344:77– 84. http://dx.doi.org/10.1042/0264-6021:3440077.

50. Darwin AJ, Ziegelhoffer EC, Kiley PJ, Stewart V. 1998. Fnr, NarP, andNarL regulation of Escherichia coli K-12 napF (periplasmic nitrate reduc-tase) operon transcription in vitro. J Bacteriol 180:4192– 4198.

51. Wang H, Tseng C-P, Gunsalus RP. 1999. The napF and narG nitratereductase operons in Escherichia coli are differentially expressed in re-sponse to submicromolar concentrations of nitrate but not nitrite. J Bac-teriol 181:5303–5308.

52. Bell LC, Richardson DJ, Ferguson SJ. 1990. Periplasmic and membrane-bound respiratory nitrate reductases in Thiosphaera pantotropha: theperiplasmic enzyme catalyzes the first step in aerobic denitrification. FEBSLett 265:85– 87. http://dx.doi.org/10.1016/0014-5793(90)80889-Q.

53. Sears HJ, Sawers G, Berks BC, Ferguson SJ, Richardson DJ. 2000.Control of periplasmic nitrate reductase gene expression (napEDABC)from Paracoccus pantotrophus in response to oxygen and carbon sub-strates. Microbiology 146:2977–2985.

54. Hartsock A, Shapleigh JP. 2011. Physiological roles for two periplasmicnitrate reductases in Rhodobacter sphaeroides 2.4.3 (ATCC 17025). J Bac-teriol 193:6483– 6489. http://dx.doi.org/10.1128/JB.05324-11.

55. Simpson PJL, Richardson DJ, Codd R. 2010. The periplasmic nitratereductase in Shewanella: the resolution, distribution and functional im-plications of two NAP isoforms, NapEDABC and NapDAGHB. Microbi-ology 156:302–312. http://dx.doi.org/10.1099/mic.0.034421-0.

56. Chen Y, Wang F, Xu J, Mehmood MA, Xiao X. 2011. Physiological andevolutionary studies of NAP systems in Shewanella piezotolerans WP3.ISME J 5:843– 855. http://dx.doi.org/10.1038/ismej.2010.182.

57. Villeneuve C, Martineau C, Mauffrey F, Villemur R. 2012. Completegenome sequences of Methylophaga sp. strain JAM1 and Methylophaga sp.strain JAM7. J Bacteriol 194:4126 – 4127. http://dx.doi.org/10.1128/JB.00726-12.

58. Delgado MJ, Bonnard N, Tresierra-Ayala A, Bedmar EJ, Müller P. 2003.The Bradyrhizobium japonicum napEDABC genes encoding the periplas-mic nitrate reductase are essential for nitrate respiration. Microbiology149:3395–3403. http://dx.doi.org/10.1099/mic.0.26620-0.

59. Baek SH, Shapleigh JP. 2005. Expression of nitrite and nitric oxide re-ductases in free-living and plant-associated Agrobacterium tumefaciensC58 cells. Appl Environ Microbiol 71:4427– 4436. http://dx.doi.org/10.1128/AEM.71.8.4427-4436.2005.

60. Tosques IE, Kwiatkowski AV, Shi J, Shapleigh JP. 1997. Characteriza-tion and regulation of the gene encoding nitrite reductase in Rhodobactersphaeroides 2.4.3. J Bacteriol 179:1090 –1095.

61. Velasco L, Mesa S, Delgado MJ, Bedmar EJ. 2001. Characterization ofthe nirK gene encoding the respiratory, Cu-containing nitrite reductase ofBradyrhizobium japonicum. Biochim Biophys Acta 1521:130 –134. http://dx.doi.org/10.1016/S0167-4781(01)00279-2.

62. Laratta WP, Shapleigh JP. 2003. Site-directed mutagenesis of NnrR: atranscriptional regulator of nitrite and nitric oxide reductase in Rhodobac-ter sphaeroides. FEMS Microbiol Lett 229:173–178. http://dx.doi.org/10.1016/S0378-1097(03)00821-8.

Martineau et al.

5014 aem.asm.org August 2015 Volume 81 Number 15Applied and Environmental Microbiology